
> ###############################################################
> ### Replication Files for:                                  ###
> ### Western Political Rhetoric and Radicalization           ###
> ### William O'Brochta, Margit Tavits, and Deniz Aksoy       ###
> ### Washington University in St. Louis                      ###
> ### Published in the British Journal of Political Science   ###
> ###############################################################
> 
> library(foreign)

> library(readstata13)

> library(haven)

> library(AER)
Loading required package: car
Loading required package: carData
Loading required package: lmtest
Loading required package: zoo

Attaching package: ‘zoo’

The following objects are masked from ‘package:base’:

    as.Date, as.Date.numeric

Loading required package: sandwich
Loading required package: survival

> library(interplot)
Loading required package: ggplot2
Loading required package: abind
Loading required package: arm
Loading required package: MASS
Loading required package: Matrix
Loading required package: lme4
Registered S3 methods overwritten by 'lme4':
  method                          from
  cooks.distance.influence.merMod car 
  influence.merMod                car 
  dfbeta.influence.merMod         car 
  dfbetas.influence.merMod        car 

arm (Version 1.10-1, built: 2018-4-12)

Working directory is /Users/williamobrochta/Dropbox/OBrochta_Tavits/BosniakSurvey


Attaching package: ‘arm’

The following object is masked from ‘package:car’:

    logit


> library(arm)

> library(stargazer)

Please cite as: 

 Hlavac, Marek (2018). stargazer: Well-Formatted Regression and Summary Statistics Tables.
 R package version 5.2.2. https://CRAN.R-project.org/package=stargazer 


> library(xtable)

Attaching package: ‘xtable’

The following object is masked from ‘package:arm’:

    display


> library(nnet)

> library(coefplot2)
Loading required package: coda

Attaching package: ‘coda’

The following object is masked from ‘package:arm’:

    traceplot


> #Preparing Dataset
> data<-read_dta('MergedDataset.dta')

> data$tr_IslamPositive<-ifelse(is.na(data$g2),0,1)

> data$tr_IslamNegative<-ifelse(is.na(data$g3),0,1)

> data$dv_USfavorable<-ifelse(data$g4==1,1,0)

> data$dv_Muslimidentity<-6-data$g5

> data$dv_UnderstandViolence<-6-data$g6

> #Recode NAs
> data$unempl<-ifelse(data$dm7==888997, NA, data$dm7)

> data$dm2<-ifelse(data$dm2==888999, NA, data$dm2)

> data$dm7<-ifelse(data$dm7==888997, NA, data$dm7)

> data$dm24<-ifelse(data$dm24==888999, NA, data$dm24)

> data$dm34<-ifelse(data$dm34==888998, NA,data$dm34)

> #Month variable
> data$month<-ifelse(data$april==1, 2, NA)

> data$month<-ifelse(data$may==1, 3, data$month)

> data$month<-ifelse(data$june==1, 4, data$month)

> data$month<-ifelse(data$july==1, 5, data$month)

> data$month<-ifelse(is.na(data$month), 1, data$month)

> data$month<-as.factor(data$month)

> #unemployed
> data$unempl<-ifelse(!is.na(data$unempl) & data$unempl==5, 1,0)

> #edu includes university education only
> data$edu<-ifelse(data$dm3_edu==3, 1,0)

> data$dv_UnderstandViolence<-as.factor(data$dv_UnderstandViolence)

> data$dv_Muslimidentity<-as.factor(data$dv_Muslimidentity)

> data$edu_factor<-as.factor(data$dm3_edu)

> data$dv_USfavorable<-as.factor(data$dv_USfavorable)

> data$married<-ifelse(data$dm19==1, 1,0)

> #Create control variable
> data$control<-ifelse(data$tr_IslamNegative==0 & data$tr_IslamPositive==0, 1,0)

> data$assignment<-ifelse(data$control==1, 0, NA)

> data$assignment<-ifelse(data$tr_IslamPositive==1, 1, data$assignment)

> data$assignment<-ifelse(data$tr_IslamNegative==1, 2, data$assignment)

> #SI.5: Comparison of Means
> #Muslim Identity
> t.test(as.numeric(data[data$tr_IslamNegative==1,]$dv_Muslimidentity),
+        as.numeric(data[data$control==1,]$dv_Muslimidentity))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamNegative == 1, ]$dv_Muslimidentity) and as.numeric(data[data$control == 1, ]$dv_Muslimidentity)
t = -0.79072, df = 1718.6, p-value = 0.4292
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.16933291  0.07202776
sample estimates:
mean of x mean of y 
 3.791169  3.839822 


> t.test(as.numeric(data[data$tr_IslamPositive==1,]$dv_Muslimidentity),
+        as.numeric(data[data$control==1,]$dv_Muslimidentity))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamPositive == 1, ]$dv_Muslimidentity) and as.numeric(data[data$control == 1, ]$dv_Muslimidentity)
t = 1.1003, df = 1766.4, p-value = 0.2714
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.05167373  0.18374036
sample estimates:
mean of x mean of y 
 3.905855  3.839822 


> mean(as.numeric(data[data$tr_IslamNegative==1,]$dv_Muslimidentity))
[1] 3.791169

> mean(as.numeric(data[data$control==1,]$dv_Muslimidentity))
[1] 3.839822

> mean(as.numeric(data[data$tr_IslamPositive==1,]$dv_Muslimidentity))
[1] 3.905855

> #U.S. Favorable (coded here as 1=unfavorable, 2=favorable; 
> #discussed in SI as percent favorable i.e., subtract 1 from these results)
> t.test(as.numeric(data[data$tr_IslamNegative==1,]$dv_USfavorable),
+        as.numeric(data[data$control==1,]$dv_USfavorable))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamNegative == 1, ]$dv_USfavorable) and as.numeric(data[data$control == 1, ]$dv_USfavorable)
t = 0.17746, df = 1726.3, p-value = 0.8592
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.04277825  0.05128943
sample estimates:
mean of x mean of y 
 1.472554  1.468298 


> t.test(as.numeric(data[data$tr_IslamPositive==1,]$dv_USfavorable),
+        as.numeric(data[data$control==1,]$dv_USfavorable))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamPositive == 1, ]$dv_USfavorable) and as.numeric(data[data$control == 1, ]$dv_USfavorable)
t = 2.4229, df = 1766.2, p-value = 0.0155
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 0.01096129 0.10410724
sample estimates:
mean of x mean of y 
 1.525832  1.468298 


> mean(as.numeric(data[data$tr_IslamNegative==1,]$dv_USfavorable))
[1] 1.472554

> mean(as.numeric(data[data$control==1,]$dv_USfavorable))
[1] 1.468298

> mean(as.numeric(data[data$tr_IslamPositive==1,]$dv_USfavorable))
[1] 1.525832

> #Understand Violence
> t.test(as.numeric(data[data$tr_IslamNegative==1,]$dv_UnderstandViolence),
+        as.numeric(data[data$control==1,]$dv_UnderstandViolence))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamNegative == 1, ]$dv_UnderstandViolence) and as.numeric(data[data$control == 1, ]$dv_UnderstandViolence)
t = 1.2344, df = 1722.9, p-value = 0.2172
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.04452913  0.19576585
sample estimates:
mean of x mean of y 
 1.983294  1.907675 


> t.test(as.numeric(data[data$tr_IslamPositive==1,]$dv_UnderstandViolence),
+        as.numeric(data[data$control==1,]$dv_UnderstandViolence))

	Welch Two Sample t-test

data:  as.numeric(data[data$tr_IslamPositive == 1, ]$dv_UnderstandViolence) and as.numeric(data[data$control == 1, ]$dv_UnderstandViolence)
t = 1.561, df = 1754, p-value = 0.1187
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.02485169  0.21868615
sample estimates:
mean of x mean of y 
 2.004592  1.907675 


> mean(as.numeric(data[data$tr_IslamNegative==1,]$dv_UnderstandViolence))
[1] 1.983294

> mean(as.numeric(data[data$control==1,]$dv_UnderstandViolence))
[1] 1.907675

> mean(as.numeric(data[data$tr_IslamPositive==1,]$dv_UnderstandViolence))
[1] 2.004592

> #Main Text Figure 1 panels a, b, c
> source('CoefPlot.R')

> pdf("dv_MuslimIdentity.pdf", width=8.5, height=11)

> par(mar = c(0, 12, 0, 0))

> #Figure 1: Muslim Identity
> #dv_MuslimIdentity
> coefplot2(c(0.06238,-0.04859),c(0.06063,0.06121), 
+          varnames=c('Pro-Islam', 'Anti-Islam'), xlim=c(-.2,.4), main='', 
+          mar=c(12,9,3,1), cex.var=2.5, cex.pts=2, h.axis=F)

> axis(3,cex.axis=1.7)

> dev.off()
null device 
          1 

> pdf("dv_USfavorable.pdf", width=8.5, height=11)

> #Figure 2: US Favorable
> #dv_USfavorable
> coefplot2(c(0.061467,0.006239),c(0.023685,0.023912), 
+           varnames=c('', ''), xlim=c(-.2,.4), main='', 
+           mar=c(12,9,3,1), cex.var=2.5, cex.pts=2, h.axis=F)

> axis(3,cex.axis=1.7)

> dev.off()
null device 
          1 

> pdf("dv_UnderstandViolence.pdf", width=8.5, height=11)

> #Figure 3: Approve Violence
> #dv_UnderstandViolence
> coefplot2(c(0.09761,0.07425),c(0.06179,0.06238), 
+          varnames=c('', ''), xlim=c(-.2,.4), main='', 
+          mar=c(12,9,3,1), cex.var=2.5, cex.pts=2, h.axis=F)

> axis(3,cex.axis=1.7)

> dev.off()
null device 
          1 

> #SI.2: Additional Information about the Survey
> data$immigration<-rowSums(data[,c(3:5)], na.rm=T)

> data$immigration<-as.factor(data$immigration)

> model<-polr(immigration~tr_IslamPositive+
+                  tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> #Table SI.2.1: Immigration Prime
> stargazer(model, star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:13
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{1}{c}{\textit{Dependent variable:}} \\ 
\cline{2-2} 
\\[-1.8ex] & immigration \\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & $-$0.026 \\ 
  & (0.053) \\ 
  & \\ 
 tr\_IslamNegative & $-$0.007 \\ 
  & (0.054) \\ 
  & \\ 
 dm1 & $-$0.056 \\ 
  & (0.045) \\ 
  & \\ 
 dm2 & $-$0.001 \\ 
  & (0.001) \\ 
  & \\ 
 dm3\_edu & $-$0.159$^{***}$ \\ 
  & (0.038) \\ 
  & \\ 
 unempl & 0.026 \\ 
  & (0.052) \\ 
  & \\ 
 married & $-$0.011 \\ 
  & (0.046) \\ 
  & \\ 
 april & 0.031 \\ 
  & (0.070) \\ 
  & \\ 
 may & $-$0.041 \\ 
  & (0.070) \\ 
  & \\ 
 june & $-$0.165$^{*}$ \\ 
  & (0.068) \\ 
  & \\ 
 july & $-$0.091 \\ 
  & (0.070) \\ 
  & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{1}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.3: Randomization and Balance Checks
> #Gender, age, education, employment status, marrital status
> #Individual covariates
> 
> #Randomization Check
> model0<-multinom(assignment~dm1+dm2+dm3_edu+unempl+married, data=data)
# weights:  21 (12 variable)
initial  value 2865.180849 
iter  10 value 2861.766460
final  value 2859.948445 
converged

> z <- summary(model0)$coefficients/summary(model0)$standard.errors

> p <- (1 - pnorm(abs(z), 0, 1)) * 2

> beta <- as.vector(t(coef(model0)))

> A<-rbind(c(1,0,0,0,0,0,-1,0,0,0,0,0), c(0,1,0,0,0,0,0,-1,0,0,0,0),
+          c(0,0,1,0,0,0,0,0,-1,0,0,0), c(0,0,0,1,0,0,0,0,0,-1,0,0),
+          c(0,0,0,0,1,0,0,0,0,0,-1,0), c(0,0,0,0,0,1,0,0,0,0,0,-1))

> t(A %*% beta) %*% solve(A %*% vcov(model0) %*% t(A), A %*% beta) 
         [,1]
[1,] 4.130906

> pchisq(4.130906, 5, lower=F)
[1] 0.5307269

> #Table SI.3.1: Randomization Check
> stargazer(model0, star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:13
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{2}{c}{\textit{Dependent variable:}} \\ 
\cline{2-3} 
\\[-1.8ex] & 1 & 2 \\ 
\\[-1.8ex] & (1) & (2)\\ 
\hline \\[-1.8ex] 
 dm1 & $-$0.165 & $-$0.090 \\ 
  & (0.097) & (0.099) \\ 
  & & \\ 
 dm2 & $-$0.002 & 0.002 \\ 
  & (0.003) & (0.003) \\ 
  & & \\ 
 dm3\_edu & $-$0.053 & $-$0.029 \\ 
  & (0.083) & (0.084) \\ 
  & & \\ 
 unempl & $-$0.020 & $-$0.0001 \\ 
  & (0.112) & (0.113) \\ 
  & & \\ 
 married & $-$0.144 & $-$0.117 \\ 
  & (0.099) & (0.100) \\ 
  & & \\ 
 Constant & 0.524 & 0.090 \\ 
  & (0.302) & (0.308) \\ 
  & & \\ 
\hline \\[-1.8ex] 
Akaike Inf. Crit. & 5,743.897 & 5,743.897 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{2}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #Table SI.3.2: Balance Checks
> #Tr 1: Islam Positive
> model0.1<-lm(tr_IslamPositive~dm1, data=data)

> mean(data[data$tr_IslamPositive==1,]$dm1)
[1] 1.528129

> mean(data[data$tr_IslamPositive==0,]$dm1)
[1] 1.555556

> model0.2<-lm(tr_IslamPositive~dm2, data=data)

> mean(data[data$tr_IslamPositive==1,]$dm2)
[1] 45.41332

> mean(data[data$tr_IslamPositive==0,]$dm2)
[1] 46.54807

> model0.3<-lm(tr_IslamPositive~dm3_edu, data=data)

> mean(data[data$tr_IslamPositive==1,]$dm3_edu)
[1] 1.871412

> mean(data[data$tr_IslamPositive==0,]$dm3_edu)
[1] 1.868739

> model0.4<-lm(tr_IslamPositive~unempl, data=data)

> mean(data[data$tr_IslamPositive==1,]$unempl, na.rm=T)
[1] 0.2640643

> mean(data[data$tr_IslamPositive==0,]$unempl, na.rm=T)
[1] 0.2619459

> model0.5<-lm(tr_IslamPositive~married, data=data)

> mean(data[data$tr_IslamPositive==1,]$married)
[1] 0.5774971

> mean(data[data$tr_IslamPositive==0,]$married)
[1] 0.6033391

> #Tr 2: Islam Negative
> model0.7<-lm(tr_IslamNegative~dm1, data=data)

> mean(data[data$tr_IslamNegative==1,]$dm1)
[1] 1.545346

> mean(data[data$tr_IslamNegative==0,]$dm1)
[1] 1.546893

> model0.8<-lm(tr_IslamNegative~dm2, data=data)

> mean(data[data$tr_IslamNegative==1,]$dm2)
[1] 46.90692

> mean(data[data$tr_IslamNegative==0,]$dm2)
[1] 45.81977

> model0.9<-lm(tr_IslamNegative~dm3_edu, data=data)

> mean(data[data$tr_IslamNegative==1,]$dm3_edu)
[1] 1.861575

> mean(data[data$tr_IslamNegative==0,]$dm3_edu)
[1] 1.873446

> model0.10<-lm(tr_IslamNegative~unempl, data=data)

> mean(data[data$tr_IslamNegative==1,]$unempl, na.rm=T)
[1] 0.2589499

> mean(data[data$tr_IslamNegative==0,]$unempl, na.rm=T)
[1] 0.2644068

> model0.11<-lm(tr_IslamNegative~married, data=data)

> mean(data[data$tr_IslamNegative==1,]$married)
[1] 0.5906921

> mean(data[data$tr_IslamNegative==0,]$married)
[1] 0.5966102

> #Control
> model0.13<-lm(control~dm1, data=data)

> mean(data[data$control==1,]$dm1)
[1] 1.565072

> mean(data[data$control==0,]$dm1)
[1] 1.536571

> model0.14<-lm(control~dm2, data=data)

> mean(data[data$control==1,]$dm2)
[1] 46.21357

> mean(data[data$control==0,]$dm2)
[1] 46.1457

> model0.15<-lm(control~dm3_edu, data=data)

> mean(data[data$control==1,]$dm3_edu)
[1] 1.875417

> mean(data[data$control==0,]$dm3_edu)
[1] 1.866589

> model0.16<-lm(control~unempl, data=data)

> mean(data[data$control==1,]$unempl, na.rm=T)
[1] 0.2647386

> mean(data[data$control==0,]$unempl, na.rm=T)
[1] 0.2615565

> model0.17<-lm(control~married, data=data)

> mean(data[data$control==1,]$married)
[1] 0.6151279

> mean(data[data$control==0,]$married)
[1] 0.5839672

> #Wald Test
> model0.18<-lm(tr_IslamPositive~dm1+dm2+dm3_edu+unempl+married, data=data)

> waldtest(model0.18)
Wald test

Model 1: tr_IslamPositive ~ dm1 + dm2 + dm3_edu + unempl + married
Model 2: tr_IslamPositive ~ 1
  Res.Df Df      F Pr(>F)
1   2602                 
2   2607 -5 1.0972 0.3598

> model0.19<-lm(tr_IslamNegative~dm1+dm2+dm3_edu+unempl+married, data=data)

> waldtest(model0.19)
Wald test

Model 1: tr_IslamNegative ~ dm1 + dm2 + dm3_edu + unempl + married
Model 2: tr_IslamNegative ~ 1
  Res.Df Df     F Pr(>F)
1   2602                
2   2607 -5 0.458 0.8077

> model0.20<-lm(control~dm1+dm2+dm3_edu+unempl+married, data=data)

> waldtest(model0.20)
Wald test

Model 1: control ~ dm1 + dm2 + dm3_edu + unempl + married
Model 2: control ~ 1
  Res.Df Df      F Pr(>F)
1   2602                 
2   2607 -5 0.9598 0.4411

> #SI.4: Regression Output
> model1<-polr(dv_UnderstandViolence~tr_IslamPositive+tr_IslamNegative+
+                april+may+june+july, 
+              data=data, method=c('probit'))

> data$dv_UnderstandViolenceN<-as.numeric(as.character(data$dv_UnderstandViolence))

> model1a<-lm(dv_UnderstandViolenceN~tr_IslamPositive+tr_IslamNegative+
+               april+may+june+july, data=data)

> model1.1<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  april+may+june+july, 
+                data=data, method=c('probit'))

> data$dv_MuslimidentityN<-as.numeric(as.character(data$dv_Muslimidentity))

> model1.1a<-lm(dv_MuslimidentityN~tr_IslamPositive+tr_IslamNegative+
+               april+may+june+july, data=data)

> model1.2<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                april+may+june+july, 
+              data=data, family = binomial(link = "probit"))

> data$dv_USfavorableN<-as.numeric(as.character(data$dv_USfavorable))

> model1.2a<-lm(dv_USfavorableN~tr_IslamPositive+tr_IslamNegative+
+                 april+may+june+july, data=data)

> #Table SI.4.1: Full Sample With Month Fixed Effects
> stargazer(model1.1, model1.1a, model1.2, model1.2a, model1, model1a, star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:15
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{6}{c}{\textit{Dependent variable:}} \\ 
\cline{2-7} 
\\[-1.8ex] & dv\_Muslimidentity & dv\_MuslimidentityN & dv\_USfavorable & dv\_USfavorableN & dv\_UnderstandViolence & dv\_UnderstandViolenceN \\ 
\\[-1.8ex] & \textit{ordered} & \textit{OLS} & \textit{probit} & \textit{OLS} & \textit{ordered} & \textit{OLS} \\ 
 & \textit{probit} & \textit{} & \textit{} & \textit{} & \textit{probit} & \textit{} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5) & (6)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & 0.055 & 0.062 & 0.155$^{**}$ & 0.061$^{**}$ & 0.081 & 0.098 \\ 
  & (0.052) & (0.061) & (0.060) & (0.024) & (0.055) & (0.062) \\ 
  & & & & & & \\ 
 tr\_IslamNegative & $-$0.038 & $-$0.049 & 0.016 & 0.006 & 0.072 & 0.074 \\ 
  & (0.052) & (0.061) & (0.060) & (0.024) & (0.055) & (0.062) \\ 
  & & & & & & \\ 
 april & $-$0.045 & $-$0.067 & 0.060 & 0.024 & $-$0.032 & $-$0.038 \\ 
  & (0.069) & (0.080) & (0.079) & (0.031) & (0.072) & (0.081) \\ 
  & & & & & & \\ 
 may & $-$0.116 & $-$0.122 & 0.272$^{***}$ & 0.108$^{***}$ & 0.035 & 0.024 \\ 
  & (0.068) & (0.079) & (0.079) & (0.031) & (0.072) & (0.081) \\ 
  & & & & & & \\ 
 june & $-$0.040 & $-$0.045 & 0.286$^{***}$ & 0.114$^{***}$ & $-$0.032 & $-$0.065 \\ 
  & (0.067) & (0.078) & (0.077) & (0.030) & (0.070) & (0.079) \\ 
  & & & & & & \\ 
 july & $-$0.024 & $-$0.019 & 0.120 & 0.048 & $-$0.010 & $-$0.056 \\ 
  & (0.069) & (0.080) & (0.079) & (0.031) & (0.072) & (0.081) \\ 
  & & & & & & \\ 
 Constant &  & 3.891$^{***}$ & $-$0.236$^{***}$ & 0.407$^{***}$ &  & 1.936$^{***}$ \\ 
  &  & (0.066) & (0.066) & (0.026) &  & (0.068) \\ 
  & & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 2,608 & 2,608 & 2,608 & 2,608 & 2,608 \\ 
R$^{2}$ &  & 0.002 &  & 0.011 &  & 0.002 \\ 
Adjusted R$^{2}$ &  & 0.0001 &  & 0.009 &  & $-$0.001 \\ 
Log Likelihood &  &  & $-$1,792.629 &  &  &  \\ 
Akaike Inf. Crit. &  &  & 3,599.258 &  &  &  \\ 
Residual Std. Error (df = 2601) &  & 1.274 &  & 0.498 &  & 1.299 \\ 
F Statistic (df = 6; 2601) &  & 1.057 &  & 4.843$^{***}$ &  & 0.750 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{6}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.5: Robustness Checks
> model2.1<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+              data=data, method=c('probit'))

> model2.2<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> model2.3<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married
+               +april+may+june+july, 
+               data=data, family = binomial(link = "probit"))

> #March Only
> model2.4<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$month==1,], method=c('probit'))

> model2.5<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$month==1,], method=c('probit'))

> model2.6<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$month==1,], family = binomial(link = "probit"))

> #April Only
> model2.7<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$april==1,], method=c('probit'))

> model2.8<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$april==1,], method=c('probit'))

> model2.9<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$april==1,], family = binomial(link = "probit"))

> #May Only
> model2.10<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$may==1,], method=c('probit'))

> model2.11<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$may==1,], method=c('probit'))

> model2.12<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$may==1,], family = binomial(link = "probit"))

> #June Only
> model2.13<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$june==1,], method=c('probit'))

> model2.14<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$june==1,], method=c('probit'))

> model2.15<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$june==1,], family = binomial(link = "probit"))

> #July Only
> model2.16<-polr(dv_UnderstandViolence~tr_IslamPositive+
+                tr_IslamNegative+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$july==1,], method=c('probit'))

> model2.17<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$july==1,], method=c('probit'))

> model2.18<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$july==1,], family = binomial(link = "probit"))

> #Table SI.5.1: Muslim Identity Full Sample By Month
> stargazer(model2.2, model2.5, model2.8, 
+           model2.11, model2.14, model2.17,
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:17
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{6}{c}{\textit{Dependent variable:}} \\ 
\cline{2-7} 
\\[-1.8ex] & \multicolumn{6}{c}{dv\_Muslimidentity} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5) & (6)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & 0.051 & 0.010 & 0.267$^{*}$ & $-$0.176 & $-$0.028 & 0.192 \\ 
  & (0.052) & (0.119) & (0.122) & (0.118) & (0.112) & (0.117) \\ 
  & & & & & & \\ 
 tr\_IslamNegative & $-$0.037 & $-$0.088 & $-$0.039 & $-$0.106 & $-$0.036 & 0.104 \\ 
  & (0.052) & (0.120) & (0.118) & (0.116) & (0.112) & (0.123) \\ 
  & & & & & & \\ 
 dm1 & $-$0.075 & $-$0.090 & $-$0.021 & $-$0.110 & $-$0.071 & $-$0.044 \\ 
  & (0.044) & (0.101) & (0.101) & (0.098) & (0.094) & (0.101) \\ 
  & & & & & & \\ 
 dm2 & $-$0.002 & $-$0.001 & $-$0.002 & $-$0.004 & $-$0.003 & 0.001 \\ 
  & (0.001) & (0.003) & (0.003) & (0.003) & (0.003) & (0.003) \\ 
  & & & & & & \\ 
 dm3\_edu & 0.021 & 0.043 & 0.174$^{*}$ & $-$0.163$^{*}$ & 0.033 & 0.052 \\ 
  & (0.037) & (0.080) & (0.087) & (0.080) & (0.084) & (0.092) \\ 
  & & & & & & \\ 
 unempl & 0.005 & $-$0.045 & $-$0.064 & 0.148 & 0.066 & $-$0.115 \\ 
  & (0.051) & (0.116) & (0.118) & (0.107) & (0.109) & (0.123) \\ 
  & & & & & & \\ 
 married & 0.010 & $-$0.045 & 0.065 & 0.089 & $-$0.044 & 0.026 \\ 
  & (0.045) & (0.103) & (0.105) & (0.099) & (0.095) & (0.100) \\ 
  & & & & & & \\ 
 april & $-$0.055 &  &  &  &  &  \\ 
  & (0.069) &  &  &  &  &  \\ 
  & & & & & & \\ 
 may & $-$0.126 &  &  &  &  &  \\ 
  & (0.068) &  &  &  &  &  \\ 
  & & & & & & \\ 
 june & $-$0.045 &  &  &  &  &  \\ 
  & (0.067) &  &  &  &  &  \\ 
  & & & & & & \\ 
 july & $-$0.027 &  &  &  &  &  \\ 
  & (0.069) &  &  &  &  &  \\ 
  & & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 518 & 508 & 512 & 566 & 504 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{6}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #Table SI.5.2: U.S. Favorable Full Sample By Month
> stargazer(model2.3, model2.6, model2.9,
+           model2.12, model2.15, model2.18, 
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:18
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{6}{c}{\textit{Dependent variable:}} \\ 
\cline{2-7} 
\\[-1.8ex] & \multicolumn{6}{c}{dv\_USfavorable} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5) & (6)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & 0.156$^{**}$ & 0.157 & 0.457$^{***}$ & $-$0.177 & 0.050 & 0.271$^{*}$ \\ 
  & (0.060) & (0.137) & (0.139) & (0.138) & (0.129) & (0.135) \\ 
  & & & & & & \\ 
 tr\_IslamNegative & 0.018 & 0.101 & 0.206 & $-$0.257 & 0.078 & $-$0.067 \\ 
  & (0.061) & (0.140) & (0.138) & (0.134) & (0.130) & (0.142) \\ 
  & & & & & & \\ 
 dm1 & $-$0.108$^{*}$ & $-$0.327$^{**}$ & $-$0.097 & $-$0.061 & $-$0.240$^{*}$ & 0.181 \\ 
  & (0.051) & (0.116) & (0.116) & (0.114) & (0.109) & (0.117) \\ 
  & & & & & & \\ 
 dm2 & $-$0.001 & $-$0.002 & $-$0.0003 & 0.001 & $-$0.001 & $-$0.004 \\ 
  & (0.001) & (0.004) & (0.004) & (0.003) & (0.003) & (0.003) \\ 
  & & & & & & \\ 
 dm3\_edu & $-$0.0001 & 0.022 & $-$0.047 & $-$0.011 & 0.062 & $-$0.046 \\ 
  & (0.043) & (0.093) & (0.099) & (0.093) & (0.097) & (0.106) \\ 
  & & & & & & \\ 
 unempl & $-$0.068 & $-$0.177 & 0.083 & $-$0.165 & $-$0.211 & 0.203 \\ 
  & (0.058) & (0.134) & (0.135) & (0.123) & (0.125) & (0.143) \\ 
  & & & & & & \\ 
 married & 0.123$^{*}$ & 0.309$^{*}$ & 0.013 & 0.102 & 0.137 & 0.091 \\ 
  & (0.051) & (0.120) & (0.121) & (0.115) & (0.110) & (0.116) \\ 
  & & & & & & \\ 
 april & 0.047 &  &  &  &  &  \\ 
  & (0.079) &  &  &  &  &  \\ 
  & & & & & & \\ 
 may & 0.273$^{***}$ &  &  &  &  &  \\ 
  & (0.079) &  &  &  &  &  \\ 
  & & & & & & \\ 
 june & 0.292$^{***}$ &  &  &  &  &  \\ 
  & (0.077) &  &  &  &  &  \\ 
  & & & & & & \\ 
 july & 0.122 &  &  &  &  &  \\ 
  & (0.079) &  &  &  &  &  \\ 
  & & & & & & \\ 
 Constant & $-$0.056 & 0.160 & $-$0.119 & 0.312 & 0.353 & $-$0.238 \\ 
  & (0.173) & (0.385) & (0.365) & (0.347) & (0.352) & (0.385) \\ 
  & & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 518 & 508 & 512 & 566 & 504 \\ 
Log Likelihood & $-$1,786.659 & $-$344.314 & $-$343.526 & $-$350.140 & $-$385.239 & $-$341.642 \\ 
Akaike Inf. Crit. & 3,597.319 & 704.627 & 703.051 & 716.280 & 786.478 & 699.284 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{6}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #Table SI.5.3: Approve Violence Full Sample By Month
> stargazer(model2.1, model2.4, model2.7,
+           model2.10, model2.13, model2.16, 
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:19
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{6}{c}{\textit{Dependent variable:}} \\ 
\cline{2-7} 
\\[-1.8ex] & \multicolumn{6}{c}{dv\_UnderstandViolence} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5) & (6)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & 0.078 & 0.105 & 0.056 & $-$0.009 & 0.095 & 0.074 \\ 
  & (0.055) & (0.126) & (0.132) & (0.124) & (0.119) & (0.121) \\ 
  & & & & & & \\ 
 tr\_IslamNegative & 0.071 & 0.108 & 0.293$^{*}$ & $-$0.123 & 0.113 & $-$0.043 \\ 
  & (0.055) & (0.128) & (0.127) & (0.122) & (0.119) & (0.129) \\ 
  & & & & & & \\ 
 dm1 & $-$0.116$^{*}$ & $-$0.162 & 0.002 & $-$0.208$^{*}$ & $-$0.043 & $-$0.156 \\ 
  & (0.046) & (0.106) & (0.108) & (0.103) & (0.100) & (0.106) \\ 
  & & & & & & \\ 
 dm2 & $-$0.002 & 0.003 & 0.003 & $-$0.003 & $-$0.004 & $-$0.005 \\ 
  & (0.001) & (0.003) & (0.003) & (0.003) & (0.003) & (0.003) \\ 
  & & & & & & \\ 
 dm3\_edu & $-$0.139$^{***}$ & $-$0.104 & $-$0.239$^{*}$ & $-$0.027 & $-$0.292$^{**}$ & $-$0.054 \\ 
  & (0.040) & (0.085) & (0.094) & (0.085) & (0.092) & (0.096) \\ 
  & & & & & & \\ 
 unempl & 0.054 & 0.086 & $-$0.019 & $-$0.010 & 0.159 & 0.125 \\ 
  & (0.053) & (0.122) & (0.127) & (0.112) & (0.113) & (0.127) \\ 
  & & & & & & \\ 
 married & 0.073 & 0.269$^{*}$ & $-$0.001 & 0.181 & $-$0.155 & 0.035 \\ 
  & (0.047) & (0.110) & (0.112) & (0.105) & (0.101) & (0.104) \\ 
  & & & & & & \\ 
 april & $-$0.041 &  &  &  &  &  \\ 
  & (0.073) &  &  &  &  &  \\ 
  & & & & & & \\ 
 may & 0.027 &  &  &  &  &  \\ 
  & (0.072) &  &  &  &  &  \\ 
  & & & & & & \\ 
 june & $-$0.036 &  &  &  &  &  \\ 
  & (0.071) &  &  &  &  &  \\ 
  & & & & & & \\ 
 july & $-$0.009 &  &  &  &  &  \\ 
  & (0.072) &  &  &  &  &  \\ 
  & & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 518 & 508 & 512 & 566 & 504 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{6}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.6: Observational Analysis
> #Hegerogeneous Treatment Effects
> #Gender
> model4.1<-polr(dv_UnderstandViolence~tr_IslamPositive*dm1+
+                  tr_IslamNegative*dm1+dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.1)

Re-fitting to get Hessian


z test of coefficients:

                       Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive     -0.0607623  0.1781651 -0.3410 0.7330697    
dm1                  -0.2006852  0.0786665 -2.5511 0.0107388 *  
tr_IslamNegative     -0.1884755  0.1813534 -1.0393 0.2986783    
dm2                  -0.0020265  0.0013648 -1.4848 0.1375927    
dm3_edu              -0.1405265  0.0399364 -3.5188 0.0004336 ***
unempl                0.0547434  0.0531348  1.0303 0.3028818    
married               0.0734251  0.0470585  1.5603 0.1186904    
april                -0.0414571  0.0727821 -0.5696 0.5689453    
may                   0.0261522  0.0719212  0.3636 0.7161391    
june                 -0.0376494  0.0706265 -0.5331 0.5939803    
july                 -0.0136805  0.0723214 -0.1892 0.8499650    
tr_IslamPositive:dm1  0.0892001  0.1104036  0.8079 0.4191223    
dm1:tr_IslamNegative  0.1677958  0.1114979  1.5049 0.1323438    
1|2                  -0.3999499  0.1896514 -2.1089 0.0349559 *  
2|3                  -0.0534854  0.1895577 -0.2822 0.7778218    
3|4                   0.3914841  0.1896734  2.0640 0.0390186 *  
4|5                   0.9756783  0.1909372  5.1099 3.223e-07 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.2<-polr(dv_Muslimidentity~tr_IslamPositive*dm1+tr_IslamNegative*dm1+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.2)

Re-fitting to get Hessian


z test of coefficients:

                       Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive     -0.1390028  0.1706422 -0.8146 0.4153093    
dm1                  -0.0999391  0.0740630 -1.3494 0.1772149    
tr_IslamNegative      0.0363495  0.1729577  0.2102 0.8335397    
dm2                  -0.0017125  0.0012893 -1.3282 0.1841007    
dm3_edu               0.0243423  0.0374131  0.6506 0.5152815    
unempl                0.0016504  0.0507233  0.0325 0.9740430    
married               0.0103641  0.0445428  0.2327 0.8160130    
april                -0.0552916  0.0689106 -0.8024 0.4223406    
may                  -0.1268010  0.0682410 -1.8581 0.0631499 .  
june                 -0.0458024  0.0669945 -0.6837 0.4941811    
july                 -0.0272199  0.0690025 -0.3945 0.6932288    
tr_IslamPositive:dm1  0.1237981  0.1049961  1.1791 0.2383690    
dm1:tr_IslamNegative -0.0477160  0.1057391 -0.4513 0.6518008    
1|2                  -1.6170076  0.1810431 -8.9316 < 2.2e-16 ***
2|3                  -1.2358129  0.1795855 -6.8815 5.924e-12 ***
3|4                  -0.6534354  0.1786075 -3.6585 0.0002537 ***
4|5                  -0.0310534  0.1785093 -0.1740 0.8618974    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.13<-glm(dv_USfavorable~tr_IslamPositive*dm1+tr_IslamNegative*dm1+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, family = binomial(link = "probit"))

> coeftest(model4.13)

z test of coefficients:

                       Estimate Std. Error z value  Pr(>|z|)    
(Intercept)          -0.1447218  0.2052478 -0.7051 0.4807434    
tr_IslamPositive      0.3483198  0.1959755  1.7774 0.0755084 .  
dm1                  -0.0506263  0.0853382 -0.5932 0.5530185    
tr_IslamNegative      0.0942903  0.1989032  0.4741 0.6354632    
dm2                  -0.0014435  0.0014845 -0.9724 0.3308615    
dm3_edu              -0.0017549  0.0431087 -0.0407 0.9675288    
unempl               -0.0657714  0.0583320 -1.1275 0.2595163    
married               0.1225528  0.0513854  2.3850 0.0170803 *  
april                 0.0472062  0.0790983  0.5968 0.5506385    
may                   0.2729459  0.0788871  3.4600 0.0005403 ***
june                  0.2929991  0.0770166  3.8044 0.0001422 ***
july                  0.1240744  0.0792021  1.5666 0.1172192    
tr_IslamPositive:dm1 -0.1247476  0.1205955 -1.0344 0.3009351    
dm1:tr_IslamNegative -0.0486745  0.1219290 -0.3992 0.6897433    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> #Age
> model4.3<-polr(dv_UnderstandViolence~tr_IslamPositive*dm2+
+                  tr_IslamNegative*dm2+dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.3)

Re-fitting to get Hessian


z test of coefficients:

                       Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive     -0.0377775  0.1476980 -0.2558 0.7981243    
dm2                  -0.0035845  0.0022424 -1.5985 0.1099291    
tr_IslamNegative     -0.0387395  0.1536076 -0.2522 0.8008883    
dm1                  -0.1174292  0.0463814 -2.5318 0.0113473 *  
dm3_edu              -0.1405978  0.0399222 -3.5218 0.0004286 ***
unempl                0.0548183  0.0530939  1.0325 0.3018483    
married               0.0736600  0.0470587  1.5653 0.1175174    
april                -0.0417492  0.0728198 -0.5733 0.5664267    
may                   0.0260006  0.0719496  0.3614 0.7178214    
june                 -0.0367902  0.0706178 -0.5210 0.6023832    
july                 -0.0086506  0.0722725 -0.1197 0.9047258    
tr_IslamPositive:dm2  0.0025104  0.0029917  0.8391 0.4014180    
dm2:tr_IslamNegative  0.0023754  0.0030795  0.7714 0.4404897    
1|2                  -0.3414134  0.1824079 -1.8717 0.0612477 .  
2|3                   0.0048410  0.1823654  0.0265 0.9788223    
3|4                   0.4497524  0.1824960  2.4645 0.0137224 *  
4|5                   1.0340817  0.1837712  5.6270 1.834e-08 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.4<-polr(dv_Muslimidentity~tr_IslamPositive*dm2+tr_IslamNegative*dm2+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.4)

Re-fitting to get Hessian


z test of coefficients:

                        Estimate  Std. Error z value  Pr(>|z|)    
tr_IslamPositive      2.7583e-01  1.3995e-01  1.9709  0.048740 *  
dm2                   8.3748e-05  2.0954e-03  0.0400  0.968118    
tr_IslamNegative     -2.5115e-02  1.4507e-01 -0.1731  0.862552    
dm1                  -7.3912e-02  4.4009e-02 -1.6795  0.093057 .  
dm3_edu               2.4803e-02  3.7422e-02  0.6628  0.507471    
unempl                3.0140e-03  5.0691e-02  0.0595  0.952586    
married               9.2585e-03  4.4550e-02  0.2078  0.835366    
april                -5.2703e-02  6.8921e-02 -0.7647  0.444457    
may                  -1.2774e-01  6.8292e-02 -1.8705  0.061417 .  
june                 -4.5867e-02  6.6999e-02 -0.6846  0.493604    
july                 -3.0352e-02  6.8943e-02 -0.4403  0.659756    
tr_IslamPositive:dm2 -4.9005e-03  2.8265e-03 -1.7338  0.082956 .  
dm2:tr_IslamNegative -2.8829e-04  2.9074e-03 -0.0992  0.921013    
1|2                  -1.4930e+00  1.7423e-01 -8.5696 < 2.2e-16 ***
2|3                  -1.1121e+00  1.7276e-01 -6.4373 1.216e-10 ***
3|4                  -5.2973e-01  1.7176e-01 -3.0841  0.002042 ** 
4|5                   9.2890e-02  1.7176e-01  0.5408  0.588638    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.14<-glm(dv_USfavorable~tr_IslamPositive*dm2+tr_IslamNegative*dm2+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, family = binomial(link = "probit"))

> coeftest(model4.14)

z test of coefficients:

                       Estimate Std. Error z value  Pr(>|z|)    
(Intercept)           0.0629961  0.1979326  0.3183 0.7502796    
tr_IslamPositive      0.0795965  0.1606395  0.4955 0.6202489    
dm2                  -0.0038014  0.0024243 -1.5681 0.1168602    
tr_IslamNegative     -0.2435368  0.1679587 -1.4500 0.1470640    
dm1                  -0.1127840  0.0507040 -2.2244 0.0261241 *  
dm3_edu              -0.0013172  0.0431244 -0.0305 0.9756328    
unempl               -0.0689007  0.0583131 -1.1816 0.2373785    
married               0.1229448  0.0514042  2.3917 0.0167694 *  
april                 0.0479010  0.0791286  0.6054 0.5449428    
may                   0.2683653  0.0789474  3.3993 0.0006756 ***
june                  0.2904500  0.0770422  3.7700 0.0001632 ***
july                  0.1211208  0.0791733  1.5298 0.1260616    
tr_IslamPositive:dm2  0.0016272  0.0032520  0.5004 0.6168133    
dm2:tr_IslamNegative  0.0056075  0.0033631  1.6674 0.0954382 .  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> #Married
> model4.7<-polr(dv_UnderstandViolence~tr_IslamPositive*married+
+                  tr_IslamNegative*married+dm3_edu+unempl+dm1+dm2
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.7)

Re-fitting to get Hessian


z test of coefficients:

                           Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive          0.0564861  0.0876437  0.6445 0.5192530    
married                   0.1124482  0.0805121  1.3967 0.1625153    
tr_IslamNegative          0.1659785  0.0882053  1.8817 0.0598728 .  
dm3_edu                  -0.1395367  0.0398851 -3.4985 0.0004679 ***
unempl                    0.0551792  0.0530755  1.0396 0.2985089    
dm1                      -0.1153227  0.0463188 -2.4898 0.0127830 *  
dm2                      -0.0019418  0.0013634 -1.4242 0.1543778    
april                    -0.0376924  0.0728403 -0.5175 0.6048310    
may                       0.0300801  0.0719279  0.4182 0.6758025    
june                     -0.0317315  0.0706971 -0.4488 0.6535487    
july                     -0.0031721  0.0723481 -0.0438 0.9650283    
tr_IslamPositive:married  0.0382721  0.1126592  0.3397 0.7340708    
married:tr_IslamNegative -0.1577580  0.1133270 -1.3921 0.1639044    
1|2                      -0.2326031  0.1646450 -1.4128 0.1577276    
2|3                       0.1137398  0.1645830  0.6911 0.4895162    
3|4                       0.5590519  0.1648118  3.3921 0.0006937 ***
4|5                       1.1441729  0.1665321  6.8706 6.394e-12 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.8<-polr(dv_Muslimidentity~tr_IslamPositive*married+
+                  tr_IslamNegative*married+dm3_edu+unempl+dm1+dm2
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.8)

Re-fitting to get Hessian


z test of coefficients:

                           Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive          0.1012303  0.0819036  1.2360 0.2164698    
married                   0.0670419  0.0753089  0.8902 0.3733449    
tr_IslamNegative          0.0164807  0.0831060  0.1983 0.8428027    
dm3_edu                   0.0216099  0.0373706  0.5783 0.5630901    
unempl                    0.0055727  0.0506738  0.1100 0.9124323    
dm1                      -0.0747443  0.0439364 -1.7012 0.0889069 .  
dm2                      -0.0017482  0.0012878 -1.3576 0.1745978    
april                    -0.0564254  0.0689539 -0.8183 0.4131828    
may                      -0.1261587  0.0682565 -1.8483 0.0645583 .  
june                     -0.0465931  0.0670669 -0.6947 0.4872275    
july                     -0.0275801  0.0690097 -0.3997 0.6894105    
tr_IslamPositive:married -0.0833072  0.1063139 -0.7836 0.4332768    
married:tr_IslamNegative -0.0884759  0.1070478 -0.8265 0.4085154    
1|2                      -1.5481536  0.1576322 -9.8213 < 2.2e-16 ***
2|3                      -1.1672470  0.1560416 -7.4804 7.412e-14 ***
3|4                      -0.5853026  0.1548992 -3.7786 0.0001577 ***
4|5                       0.0368106  0.1548118  0.2378 0.8120542    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.16<-glm(dv_USfavorable~tr_IslamPositive*married+tr_IslamNegative*married+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, family = binomial(link = "probit"))

> coeftest(model4.16)

z test of coefficients:

                            Estimate  Std. Error z value  Pr(>|z|)    
(Intercept)              -0.06935515  0.17806325 -0.3895 0.6969083    
tr_IslamPositive          0.14731224  0.09450619  1.5588 0.1190538    
married                   0.14128149  0.08712484  1.6216 0.1048894    
tr_IslamNegative          0.06183621  0.09618747  0.6429 0.5203073    
dm1                      -0.10763531  0.05059917 -2.1272 0.0334022 *  
dm2                      -0.00146893  0.00148226 -0.9910 0.3216820    
dm3_edu                  -0.00025898  0.04305620 -0.0060 0.9952007    
unempl                   -0.06810549  0.05828184 -1.1686 0.2425833    
april                     0.04894787  0.07916618  0.6183 0.5363824    
may                       0.27445440  0.07891194  3.4780 0.0005052 ***
june                      0.29446049  0.07712836  3.8178 0.0001346 ***
july                      0.12503949  0.07925000  1.5778 0.1146149    
tr_IslamPositive:married  0.01566842  0.12237466  0.1280 0.8981201    
married:tr_IslamNegative -0.07307225  0.12370911 -0.5907 0.5547362    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> #Unemployed
> model4.9<-polr(dv_UnderstandViolence~tr_IslamPositive*unempl+
+                  tr_IslamNegative*unempl+dm3_edu+unempl+married+dm1+dm2
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.9)

Re-fitting to get Hessian


z test of coefficients:

                           Estimate  Std. Error z value  Pr(>|z|)    
tr_IslamPositive         0.07830445  0.06441316  1.2157 0.2241147    
unempl                   0.05545077  0.08840376  0.6272 0.5304990    
tr_IslamNegative         0.07153438  0.06466484  1.1062 0.2686257    
dm3_edu                 -0.13903549  0.03987808 -3.4865 0.0004894 ***
married                  0.07316201  0.04705132  1.5549 0.1199602    
dm1                     -0.11554915  0.04636804 -2.4920 0.0127026 *  
dm2                     -0.00192924  0.00136393 -1.4145 0.1572219    
april                   -0.04110618  0.07281567 -0.5645 0.5723977    
may                      0.02723085  0.07190624  0.3787 0.7049111    
june                    -0.03652160  0.07064980 -0.5169 0.6051992    
july                    -0.00922324  0.07228864 -0.1276 0.8984741    
tr_IslamPositive:unempl -0.00273924  0.12405726 -0.0221 0.9823838    
unempl:tr_IslamNegative -0.00038779  0.12523490 -0.0031 0.9975294    
1|2                     -0.25892756  0.16079339 -1.6103 0.1073297    
2|3                      0.08731575  0.16072982  0.5432 0.5869608    
3|4                      0.53211882  0.16094167  3.3063 0.0009454 ***
4|5                      1.11607877  0.16261379  6.8634 6.725e-12 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.10<-polr(dv_Muslimidentity~tr_IslamPositive*unempl+tr_IslamNegative*unempl+
+                  dm3_edu+unempl+married+dm1+dm2
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.10)

Re-fitting to get Hessian


z test of coefficients:

                          Estimate Std. Error  z value  Pr(>|z|)    
tr_IslamPositive         0.0729618  0.0608877   1.1983   0.23080    
unempl                   0.0241156  0.0838274   0.2877   0.77359    
tr_IslamNegative        -0.0434185  0.0608907  -0.7131   0.47581    
dm3_edu                  0.0215855  0.0373700   0.5776   0.56352    
married                  0.0106235  0.0445411   0.2385   0.81149    
dm1                     -0.0729324  0.0439805  -1.6583   0.09726 .  
dm2                     -0.0017826  0.0012885  -1.3835   0.16651    
april                   -0.0565071  0.0689186  -0.8199   0.41227    
may                     -0.1265660  0.0682435  -1.8546   0.06365 .  
june                    -0.0472020  0.0670225  -0.7043   0.48126    
july                    -0.0297541  0.0689715  -0.4314   0.66618    
tr_IslamPositive:unempl -0.0819528  0.1183374  -0.6925   0.48860    
unempl:tr_IslamNegative  0.0252330  0.1195734   0.2110   0.83287    
1|2                     -1.5771932  0.1544576 -10.2112 < 2.2e-16 ***
2|3                     -1.1965635  0.1527823  -7.8318 4.809e-15 ***
3|4                     -0.6145674  0.1515877  -4.0542 5.031e-05 ***
4|5                      0.0077230  0.1514960   0.0510   0.95934    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.17<-glm(dv_USfavorable~tr_IslamPositive*unempl+tr_IslamNegative*unempl+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, family = binomial(link = "probit"))

> coeftest(model4.17)

z test of coefficients:

                           Estimate  Std. Error z value  Pr(>|z|)    
(Intercept)             -6.3931e-02  1.7407e-01 -0.3673 0.7134140    
tr_IslamPositive         1.7788e-01  6.9980e-02  2.5419 0.0110255 *  
unempl                  -3.9162e-02  9.6388e-02 -0.4063 0.6845249    
tr_IslamNegative         1.9139e-02  7.0421e-02  0.2718 0.7857937    
dm1                     -1.0659e-01  5.0638e-02 -2.1049 0.0352978 *  
dm2                     -1.4967e-03  1.4834e-03 -1.0090 0.3129888    
dm3_edu                 -6.9991e-05  4.3057e-02 -0.0016 0.9987030    
married                  1.2308e-01  5.1386e-02  2.3952 0.0166131 *  
april                    4.6461e-02  7.9101e-02  0.5874 0.5569643    
may                      2.7269e-01  7.8879e-02  3.4571 0.0005461 ***
june                     2.9081e-01  7.7039e-02  3.7748 0.0001601 ***
july                     1.2036e-01  7.9177e-02  1.5201 0.1284734    
tr_IslamPositive:unempl -8.3854e-02  1.3588e-01 -0.6171 0.5371447    
unempl:tr_IslamNegative -3.3820e-03  1.3760e-01 -0.0246 0.9803911    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> #Educated
> model4.11<-polr(dv_UnderstandViolence~tr_IslamPositive*dm3_edu+
+                  tr_IslamNegative*dm3_edu+dm3_edu+unempl+married+dm1+dm2
+                +april+may+june+july, 
+                data=data, method=c('probit'))

> coeftest(model4.11)

Re-fitting to get Hessian


z test of coefficients:

                           Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive          0.1846922  0.1756565  1.0514 0.2930565    
dm3_edu                  -0.0718602  0.0648447 -1.1082 0.2677799    
tr_IslamNegative          0.3526955  0.1786004  1.9748 0.0482937 *  
unempl                    0.0549653  0.0530855  1.0354 0.3004771    
married                   0.0732313  0.0470615  1.5561 0.1196899    
dm1                      -0.1149671  0.0463459 -2.4806 0.0131151 *  
dm2                      -0.0019243  0.0013638 -1.4110 0.1582436    
april                    -0.0428683  0.0728117 -0.5888 0.5560249    
may                       0.0253094  0.0719294  0.3519 0.7249396    
june                     -0.0393132  0.0706383 -0.5565 0.5778406    
july                     -0.0094260  0.0722571 -0.1305 0.8962093    
tr_IslamPositive:dm3_edu -0.0575095  0.0895615 -0.6421 0.5207933    
dm3_edu:tr_IslamNegative -0.1520995  0.0917777 -1.6573 0.0974669 .  
1|2                      -0.1343426  0.1856512 -0.7236 0.4692936    
2|3                       0.2120978  0.1856054  1.1427 0.2531485    
3|4                       0.6571227  0.1858385  3.5360 0.0004063 ***
4|5                       1.2416313  0.1875145  6.6215 3.555e-11 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.12<-polr(dv_Muslimidentity~tr_IslamPositive*dm3_edu+tr_IslamNegative*dm3_edu+
+                   dm3_edu+unempl+married+dm1+dm2
+                 +april+may+june+july, 
+                 data=data, method=c('probit'))

> coeftest(model4.12)

Re-fitting to get Hessian


z test of coefficients:

                           Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive         -0.0387412  0.1660075 -0.2334   0.81547    
dm3_edu                  -0.0227878  0.0600152 -0.3797   0.70417    
tr_IslamNegative         -0.2079391  0.1689250 -1.2310   0.21834    
unempl                    0.0048612  0.0506806  0.0959   0.92359    
married                   0.0103711  0.0445417  0.2328   0.81589    
dm1                      -0.0741117  0.0439752 -1.6853   0.09193 .  
dm2                      -0.0017521  0.0012884 -1.3599   0.17387    
april                    -0.0538834  0.0689230 -0.7818   0.43434    
may                      -0.1254617  0.0682437 -1.8384   0.06600 .  
june                     -0.0431735  0.0670055 -0.6443   0.51936    
july                     -0.0272375  0.0689288 -0.3952   0.69273    
tr_IslamPositive:dm3_edu  0.0479892  0.0842770  0.5694   0.56907    
dm3_edu:tr_IslamNegative  0.0915732  0.0860817  1.0638   0.28742    
1|2                      -1.6642024  0.1762685 -9.4413 < 2.2e-16 ***
2|3                      -1.2834158  0.1748017 -7.3421 2.102e-13 ***
3|4                      -0.7012717  0.1736875 -4.0375 5.401e-05 ***
4|5                      -0.0790198  0.1735262 -0.4554   0.64884    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model4.18<-glm(dv_USfavorable~tr_IslamPositive*dm3_edu+tr_IslamNegative*dm3_edu+
+                  dm1+dm2+dm3_edu+unempl+married
+                +april+may+june+july, 
+                data=data, family = binomial(link = "probit"))

> coeftest(model4.18)

z test of coefficients:

                           Estimate Std. Error z value  Pr(>|z|)    
(Intercept)              -0.2091166  0.2007771 -1.0415 0.2976269    
tr_IslamPositive          0.3827199  0.1914252  1.9993 0.0455739 *  
dm3_edu                   0.0817708  0.0697810  1.1718 0.2412687    
tr_IslamNegative          0.2628510  0.1958654  1.3420 0.1795966    
dm1                      -0.1098349  0.0506491 -2.1685 0.0301170 *  
dm2                      -0.0014093  0.0014836 -0.9499 0.3421478    
unempl                   -0.0664669  0.0583022 -1.1400 0.2542690    
married                   0.1232286  0.0514001  2.3974 0.0165102 *  
april                     0.0452275  0.0791235  0.5716 0.5675886    
may                       0.2713623  0.0789121  3.4388 0.0005843 ***
june                      0.2903478  0.0770535  3.7681 0.0001645 ***
july                      0.1222604  0.0791460  1.5447 0.1224080    
tr_IslamPositive:dm3_edu -0.1211801  0.0969876 -1.2494 0.2115047    
dm3_edu:tr_IslamNegative -0.1308902  0.0996959 -1.3129 0.1892183    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> #SI.6.1
> stargazer(model4.2, model4.4, 
+           model4.8, model4.10, model4.12, 
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:24
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{5}{c}{\textit{Dependent variable:}} \\ 
\cline{2-6} 
\\[-1.8ex] & \multicolumn{5}{c}{dv\_Muslimidentity} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & $-$0.139 & 0.276$^{*}$ & 0.101 & 0.073 & $-$0.039 \\ 
  & (0.171) & (0.140) & (0.082) & (0.061) & (0.166) \\ 
  & & & & & \\ 
 dm1 & $-$0.100 & $-$0.074 & $-$0.075 & $-$0.073 & $-$0.074 \\ 
  & (0.074) & (0.044) & (0.044) & (0.044) & (0.044) \\ 
  & & & & & \\ 
 tr\_IslamNegative & 0.036 & $-$0.025 & 0.016 & $-$0.043 & $-$0.208 \\ 
  & (0.173) & (0.145) & (0.083) & (0.061) & (0.169) \\ 
  & & & & & \\ 
 dm2 & $-$0.002 & 0.0001 & $-$0.002 & $-$0.002 & $-$0.002 \\ 
  & (0.001) & (0.002) & (0.001) & (0.001) & (0.001) \\ 
  & & & & & \\ 
 dm3\_edu & 0.024 & 0.025 & 0.022 & 0.022 & $-$0.023 \\ 
  & (0.037) & (0.037) & (0.037) & (0.037) & (0.060) \\ 
  & & & & & \\ 
 unempl & 0.002 & 0.003 & 0.006 & 0.024 & 0.005 \\ 
  & (0.051) & (0.051) & (0.051) & (0.084) & (0.051) \\ 
  & & & & & \\ 
 married & 0.010 & 0.009 & 0.067 & 0.011 & 0.010 \\ 
  & (0.045) & (0.045) & (0.075) & (0.045) & (0.045) \\ 
  & & & & & \\ 
 april & $-$0.055 & $-$0.053 & $-$0.056 & $-$0.057 & $-$0.054 \\ 
  & (0.069) & (0.069) & (0.069) & (0.069) & (0.069) \\ 
  & & & & & \\ 
 may & $-$0.127 & $-$0.128 & $-$0.126 & $-$0.127 & $-$0.125 \\ 
  & (0.068) & (0.068) & (0.068) & (0.068) & (0.068) \\ 
  & & & & & \\ 
 june & $-$0.046 & $-$0.046 & $-$0.047 & $-$0.047 & $-$0.043 \\ 
  & (0.067) & (0.067) & (0.067) & (0.067) & (0.067) \\ 
  & & & & & \\ 
 july & $-$0.027 & $-$0.030 & $-$0.028 & $-$0.030 & $-$0.027 \\ 
  & (0.069) & (0.069) & (0.069) & (0.069) & (0.069) \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm1 & 0.124 &  &  &  &  \\ 
  & (0.105) &  &  &  &  \\ 
  & & & & & \\ 
 dm1:tr\_IslamNegative & $-$0.048 &  &  &  &  \\ 
  & (0.106) &  &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm2 &  & $-$0.005 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 dm2:tr\_IslamNegative &  & $-$0.0003 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:married &  &  & $-$0.083 &  &  \\ 
  &  &  & (0.106) &  &  \\ 
  & & & & & \\ 
 married:tr\_IslamNegative &  &  & $-$0.088 &  &  \\ 
  &  &  & (0.107) &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:unempl &  &  &  & $-$0.082 &  \\ 
  &  &  &  & (0.118) &  \\ 
  & & & & & \\ 
 unempl:tr\_IslamNegative &  &  &  & 0.025 &  \\ 
  &  &  &  & (0.120) &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm3\_edu &  &  &  &  & 0.048 \\ 
  &  &  &  &  & (0.084) \\ 
  & & & & & \\ 
 dm3\_edu:tr\_IslamNegative &  &  &  &  & 0.092 \\ 
  &  &  &  &  & (0.086) \\ 
  & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 2,608 & 2,608 & 2,608 & 2,608 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{5}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.6.2
> stargazer(model4.13, model4.14,
+           model4.16, model4.17, model4.18, 
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:25
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{5}{c}{\textit{Dependent variable:}} \\ 
\cline{2-6} 
\\[-1.8ex] & \multicolumn{5}{c}{dv\_USfavorable} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & 0.348 & 0.080 & 0.147 & 0.178$^{*}$ & 0.383$^{*}$ \\ 
  & (0.196) & (0.161) & (0.095) & (0.070) & (0.191) \\ 
  & & & & & \\ 
 dm1 & $-$0.051 & $-$0.113$^{*}$ & $-$0.108$^{*}$ & $-$0.107$^{*}$ & $-$0.110$^{*}$ \\ 
  & (0.085) & (0.051) & (0.051) & (0.051) & (0.051) \\ 
  & & & & & \\ 
 tr\_IslamNegative & 0.094 & $-$0.244 & 0.062 & 0.019 & 0.263 \\ 
  & (0.199) & (0.168) & (0.096) & (0.070) & (0.196) \\ 
  & & & & & \\ 
 dm2 & $-$0.001 & $-$0.004 & $-$0.001 & $-$0.001 & $-$0.001 \\ 
  & (0.001) & (0.002) & (0.001) & (0.001) & (0.001) \\ 
  & & & & & \\ 
 dm3\_edu & $-$0.002 & $-$0.001 & $-$0.0003 & $-$0.0001 & 0.082 \\ 
  & (0.043) & (0.043) & (0.043) & (0.043) & (0.070) \\ 
  & & & & & \\ 
 unempl & $-$0.066 & $-$0.069 & $-$0.068 & $-$0.039 & $-$0.066 \\ 
  & (0.058) & (0.058) & (0.058) & (0.096) & (0.058) \\ 
  & & & & & \\ 
 married & 0.123$^{*}$ & 0.123$^{*}$ & 0.141 & 0.123$^{*}$ & 0.123$^{*}$ \\ 
  & (0.051) & (0.051) & (0.087) & (0.051) & (0.051) \\ 
  & & & & & \\ 
 april & 0.047 & 0.048 & 0.049 & 0.046 & 0.045 \\ 
  & (0.079) & (0.079) & (0.079) & (0.079) & (0.079) \\ 
  & & & & & \\ 
 may & 0.273$^{***}$ & 0.268$^{***}$ & 0.274$^{***}$ & 0.273$^{***}$ & 0.271$^{***}$ \\ 
  & (0.079) & (0.079) & (0.079) & (0.079) & (0.079) \\ 
  & & & & & \\ 
 june & 0.293$^{***}$ & 0.290$^{***}$ & 0.294$^{***}$ & 0.291$^{***}$ & 0.290$^{***}$ \\ 
  & (0.077) & (0.077) & (0.077) & (0.077) & (0.077) \\ 
  & & & & & \\ 
 july & 0.124 & 0.121 & 0.125 & 0.120 & 0.122 \\ 
  & (0.079) & (0.079) & (0.079) & (0.079) & (0.079) \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm1 & $-$0.125 &  &  &  &  \\ 
  & (0.121) &  &  &  &  \\ 
  & & & & & \\ 
 dm1:tr\_IslamNegative & $-$0.049 &  &  &  &  \\ 
  & (0.122) &  &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm2 &  & 0.002 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 dm2:tr\_IslamNegative &  & 0.006 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:married &  &  & 0.016 &  &  \\ 
  &  &  & (0.122) &  &  \\ 
  & & & & & \\ 
 married:tr\_IslamNegative &  &  & $-$0.073 &  &  \\ 
  &  &  & (0.124) &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:unempl &  &  &  & $-$0.084 &  \\ 
  &  &  &  & (0.136) &  \\ 
  & & & & & \\ 
 unempl:tr\_IslamNegative &  &  &  & $-$0.003 &  \\ 
  &  &  &  & (0.138) &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm3\_edu &  &  &  &  & $-$0.121 \\ 
  &  &  &  &  & (0.097) \\ 
  & & & & & \\ 
 dm3\_edu:tr\_IslamNegative &  &  &  &  & $-$0.131 \\ 
  &  &  &  &  & (0.100) \\ 
  & & & & & \\ 
 Constant & $-$0.145 & 0.063 & $-$0.069 & $-$0.064 & $-$0.209 \\ 
  & (0.205) & (0.198) & (0.178) & (0.174) & (0.201) \\ 
  & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 2,608 & 2,608 & 2,608 & 2,608 \\ 
Log Likelihood & $-$1,786.117 & $-$1,785.196 & $-$1,786.369 & $-$1,786.419 & $-$1,785.548 \\ 
Akaike Inf. Crit. & 3,600.233 & 3,598.391 & 3,600.737 & 3,600.838 & 3,599.096 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{5}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.6.3
> stargazer(model4.1, model4.3, 
+           model4.7,model4.9, model4.11, 
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:27
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{5}{c}{\textit{Dependent variable:}} \\ 
\cline{2-6} 
\\[-1.8ex] & \multicolumn{5}{c}{dv\_UnderstandViolence} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & $-$0.061 & $-$0.038 & 0.056 & 0.078 & 0.185 \\ 
  & (0.178) & (0.148) & (0.088) & (0.064) & (0.176) \\ 
  & & & & & \\ 
 dm1 & $-$0.201$^{*}$ & $-$0.117$^{*}$ & $-$0.115$^{*}$ & $-$0.116$^{*}$ & $-$0.115$^{*}$ \\ 
  & (0.079) & (0.046) & (0.046) & (0.046) & (0.046) \\ 
  & & & & & \\ 
 tr\_IslamNegative & $-$0.188 & $-$0.039 & 0.166 & 0.072 & 0.353$^{*}$ \\ 
  & (0.181) & (0.154) & (0.088) & (0.065) & (0.179) \\ 
  & & & & & \\ 
 dm2 & $-$0.002 & $-$0.004 & $-$0.002 & $-$0.002 & $-$0.002 \\ 
  & (0.001) & (0.002) & (0.001) & (0.001) & (0.001) \\ 
  & & & & & \\ 
 dm3\_edu & $-$0.141$^{***}$ & $-$0.141$^{***}$ & $-$0.140$^{***}$ & $-$0.139$^{***}$ & $-$0.072 \\ 
  & (0.040) & (0.040) & (0.040) & (0.040) & (0.065) \\ 
  & & & & & \\ 
 unempl & 0.055 & 0.055 & 0.055 & 0.055 & 0.055 \\ 
  & (0.053) & (0.053) & (0.053) & (0.088) & (0.053) \\ 
  & & & & & \\ 
 married & 0.073 & 0.074 & 0.112 & 0.073 & 0.073 \\ 
  & (0.047) & (0.047) & (0.081) & (0.047) & (0.047) \\ 
  & & & & & \\ 
 april & $-$0.041 & $-$0.042 & $-$0.038 & $-$0.041 & $-$0.043 \\ 
  & (0.073) & (0.073) & (0.073) & (0.073) & (0.073) \\ 
  & & & & & \\ 
 may & 0.026 & 0.026 & 0.030 & 0.027 & 0.025 \\ 
  & (0.072) & (0.072) & (0.072) & (0.072) & (0.072) \\ 
  & & & & & \\ 
 june & $-$0.038 & $-$0.037 & $-$0.032 & $-$0.037 & $-$0.039 \\ 
  & (0.071) & (0.071) & (0.071) & (0.071) & (0.071) \\ 
  & & & & & \\ 
 july & $-$0.014 & $-$0.009 & $-$0.003 & $-$0.009 & $-$0.009 \\ 
  & (0.072) & (0.072) & (0.072) & (0.072) & (0.072) \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm1 & 0.089 &  &  &  &  \\ 
  & (0.110) &  &  &  &  \\ 
  & & & & & \\ 
 dm1:tr\_IslamNegative & 0.168 &  &  &  &  \\ 
  & (0.111) &  &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm2 &  & 0.003 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 dm2:tr\_IslamNegative &  & 0.002 &  &  &  \\ 
  &  & (0.003) &  &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:married &  &  & 0.038 &  &  \\ 
  &  &  & (0.113) &  &  \\ 
  & & & & & \\ 
 married:tr\_IslamNegative &  &  & $-$0.158 &  &  \\ 
  &  &  & (0.113) &  &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:unempl &  &  &  & $-$0.003 &  \\ 
  &  &  &  & (0.124) &  \\ 
  & & & & & \\ 
 unempl:tr\_IslamNegative &  &  &  & $-$0.0004 &  \\ 
  &  &  &  & (0.125) &  \\ 
  & & & & & \\ 
 tr\_IslamPositive:dm3\_edu &  &  &  &  & $-$0.058 \\ 
  &  &  &  &  & (0.090) \\ 
  & & & & & \\ 
 dm3\_edu:tr\_IslamNegative &  &  &  &  & $-$0.152 \\ 
  &  &  &  &  & (0.092) \\ 
  & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 2,608 & 2,608 & 2,608 & 2,608 & 2,608 \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{5}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 

> #SI.6.4: Region
> #1 is FBiH, 2 RS, 3 Brcko (excluded)
> data$entity_dummy<-ifelse(data$entity==1, 1, NA)

> data$entity_dummy<-ifelse(data$entity==2, 0, data$entity)

> model5<-polr(dv_UnderstandViolence~tr_IslamPositive+tr_IslamNegative+
+                 april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$entity_dummy==0,], method=c('probit'))

> coeftest(model5)

Re-fitting to get Hessian


z test of coefficients:

                   Estimate Std. Error z value Pr(>|z|)   
tr_IslamPositive  0.0340646  0.2047134  0.1664 0.867841   
tr_IslamNegative  0.2352001  0.2100751  1.1196 0.262884   
april            -0.1193350  0.3109598 -0.3838 0.701154   
may              -0.1094319  0.2746388 -0.3985 0.690293   
june             -0.0936316  0.2578554 -0.3631 0.716518   
july             -0.7420119  0.3044137 -2.4375 0.014789 * 
dm1               0.0156770  0.1842937  0.0851 0.932209   
dm2               0.0108176  0.0059108  1.8301 0.067228 . 
dm3_edu          -0.1731242  0.1515255 -1.1425 0.253229   
unempl            0.3322189  0.2243849  1.4806 0.138720   
married          -0.2036020  0.1797435 -1.1327 0.257325   
1|2               0.6397572  0.6754205  0.9472 0.343538   
2|3               0.9718261  0.6757951  1.4380 0.150420   
3|4               1.4104378  0.6778826  2.0807 0.037466 * 
4|5               1.8446420  0.6845925  2.6945 0.007049 **
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model5.1<-polr(dv_UnderstandViolence~tr_IslamPositive+tr_IslamNegative+
+                 april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$entity_dummy==1,], method=c('probit'))

> coeftest(model5.1)

Re-fitting to get Hessian


z test of coefficients:

                   Estimate Std. Error z value  Pr(>|z|)    
tr_IslamPositive  0.0763141  0.0573410  1.3309 0.1832280    
tr_IslamNegative  0.0502769  0.0576144  0.8726 0.3828569    
april            -0.0231521  0.0750443 -0.3085 0.7576924    
may               0.0514142  0.0748763  0.6867 0.4923003    
june             -0.0260501  0.0736467 -0.3537 0.7235511    
july              0.0464188  0.0747163  0.6213 0.5344240    
dm1              -0.1288284  0.0481647 -2.6747 0.0074785 ** 
dm2              -0.0022719  0.0014270 -1.5921 0.1113539    
dm3_edu          -0.1426497  0.0416597 -3.4242 0.0006167 ***
unempl            0.0577518  0.0551343  1.0475 0.2948809    
married           0.0977613  0.0490476  1.9932 0.0462401 *  
1|2              -0.2956952  0.1653600 -1.7882 0.0737453 .  
2|3               0.0545646  0.1652926  0.3301 0.7413175    
3|4               0.5031061  0.1655120  3.0397 0.0023682 ** 
4|5               1.1014422  0.1672693  6.5848 4.554e-11 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model5.2<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+              data=data[data$entity_dummy==0,], method=c('probit'))

> coeftest(model5.2)

Re-fitting to get Hessian


z test of coefficients:

                   Estimate Std. Error z value Pr(>|z|)  
tr_IslamPositive -0.2077159  0.1890662 -1.0986  0.27192  
tr_IslamNegative -0.1731743  0.2052377 -0.8438  0.39880  
april            -0.2570069  0.2936879 -0.8751  0.38152  
may              -0.3954791  0.2632138 -1.5025  0.13297  
june             -0.0076194  0.2557375 -0.0298  0.97623  
july              0.3347766  0.2817960  1.1880  0.23483  
dm1               0.3070212  0.1746164  1.7583  0.07870 .
dm2               0.0060733  0.0052978  1.1464  0.25164  
dm3_edu           0.0509176  0.1386857  0.3671  0.71351  
unempl            0.3761017  0.2089258  1.8002  0.07183 .
married           0.0945423  0.1691564  0.5589  0.57623  
1|2              -0.9093681  0.6402370 -1.4204  0.15550  
2|3              -0.6909433  0.6340517 -1.0897  0.27583  
3|4               0.1375200  0.6268048  0.2194  0.82634  
4|5               0.6925183  0.6288521  1.1012  0.27079  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model5.3<-polr(dv_Muslimidentity~tr_IslamPositive+tr_IslamNegative+
+                  april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+                data=data[data$entity_dummy==1,], method=c('probit'))

> coeftest(model5.3)

Re-fitting to get Hessian


z test of coefficients:

                   Estimate Std. Error  z value  Pr(>|z|)    
tr_IslamPositive  0.0767959  0.0544557   1.4102   0.15847    
tr_IslamNegative -0.0144858  0.0544693  -0.2659   0.79028    
april            -0.0412435  0.0710416  -0.5806   0.56154    
may              -0.1174613  0.0710562  -1.6531   0.09832 .  
june             -0.0657531  0.0696745  -0.9437   0.34531    
july             -0.0602338  0.0713872  -0.8438   0.39880    
dm1              -0.0918310  0.0457288  -2.0082   0.04463 *  
dm2              -0.0032186  0.0013477  -2.3882   0.01693 *  
dm3_edu           0.0375586  0.0390938   0.9607   0.33669    
unempl           -0.0256583  0.0526624  -0.4872   0.62610    
married           0.0187377  0.0464095   0.4037   0.68640    
1|2              -1.6183973  0.1589812 -10.1798 < 2.2e-16 ***
2|3              -1.2259793  0.1572547  -7.7961 6.383e-15 ***
3|4              -0.6515355  0.1560637  -4.1748 2.982e-05 ***
4|5              -0.0175411  0.1559107  -0.1125   0.91042    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model5.4<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$entity_dummy==0,], family = binomial(link = "probit"))

> coeftest(model5.4)

z test of coefficients:

                   Estimate Std. Error z value Pr(>|z|)  
(Intercept)       0.5928556  0.7118070  0.8329  0.40491  
tr_IslamPositive  0.3175986  0.2094104  1.5166  0.12936  
tr_IslamNegative  0.3321155  0.2239807  1.4828  0.13813  
april            -0.3858395  0.3352361 -1.1509  0.24975  
may              -0.6021594  0.3024863 -1.9907  0.04651 *
june             -0.1939130  0.2909587 -0.6665  0.50512  
july             -0.4470174  0.3055843 -1.4628  0.14351  
dm1              -0.1379346  0.1943478 -0.7097  0.47787  
dm2              -0.0054082  0.0060182 -0.8986  0.36885  
dm3_edu           0.1844033  0.1562321  1.1803  0.23787  
unempl           -0.0667216  0.2286635 -0.2918  0.77045  
married           0.0531373  0.1877420  0.2830  0.77715  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> model5.5<-glm(dv_USfavorable~tr_IslamPositive+tr_IslamNegative+
+                 april+may+june+july+dm1+dm2+dm3_edu+unempl+married, 
+               data=data[data$entity_dummy==1,], family = binomial(link = "probit"))

> coeftest(model5.5)

z test of coefficients:

                   Estimate Std. Error z value  Pr(>|z|)    
(Intercept)      -0.0744277  0.1796573 -0.4143   0.67867    
tr_IslamPositive  0.1446771  0.0628338  2.3025   0.02130 *  
tr_IslamNegative -0.0034995  0.0632083 -0.0554   0.95585    
april             0.0713274  0.0819066  0.8708   0.38384    
may               0.3341353  0.0824853  4.0508 5.103e-05 ***
june              0.3145124  0.0805095  3.9065 9.363e-05 ***
july              0.1553062  0.0824615  1.8834   0.05965 .  
dm1              -0.0974730  0.0528399 -1.8447   0.06508 .  
dm2              -0.0022095  0.0015582 -1.4180   0.15619    
dm3_edu          -0.0047383  0.0452181 -0.1048   0.91654    
unempl           -0.0760427  0.0608699 -1.2493   0.21157    
married           0.1332103  0.0538102  2.4756   0.01330 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


> stargazer(model5.2, model5.3, model5.4, model5.5, model5, model5.1,
+           star.cutoffs = c(0.05, 0.01, 0.001))

% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Mon, Jul 06, 2020 - 21:25:30
\begin{table}[!htbp] \centering 
  \caption{} 
  \label{} 
\begin{tabular}{@{\extracolsep{5pt}}lcccccc} 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
 & \multicolumn{6}{c}{\textit{Dependent variable:}} \\ 
\cline{2-7} 
\\[-1.8ex] & \multicolumn{2}{c}{dv\_Muslimidentity} & \multicolumn{2}{c}{dv\_USfavorable} & \multicolumn{2}{c}{dv\_UnderstandViolence} \\ 
\\[-1.8ex] & \multicolumn{2}{c}{\textit{ordered}} & \multicolumn{2}{c}{\textit{probit}} & \multicolumn{2}{c}{\textit{ordered}} \\ 
 & \multicolumn{2}{c}{\textit{probit}} & \multicolumn{2}{c}{\textit{}} & \multicolumn{2}{c}{\textit{probit}} \\ 
\\[-1.8ex] & (1) & (2) & (3) & (4) & (5) & (6)\\ 
\hline \\[-1.8ex] 
 tr\_IslamPositive & $-$0.208 & 0.077 & 0.318 & 0.145$^{*}$ & 0.034 & 0.076 \\ 
  & (0.189) & (0.054) & (0.209) & (0.063) & (0.205) & (0.057) \\ 
  & & & & & & \\ 
 tr\_IslamNegative & $-$0.173 & $-$0.014 & 0.332 & $-$0.003 & 0.235 & 0.050 \\ 
  & (0.205) & (0.054) & (0.224) & (0.063) & (0.210) & (0.058) \\ 
  & & & & & & \\ 
 april & $-$0.257 & $-$0.041 & $-$0.386 & 0.071 & $-$0.119 & $-$0.023 \\ 
  & (0.294) & (0.071) & (0.335) & (0.082) & (0.311) & (0.075) \\ 
  & & & & & & \\ 
 may & $-$0.395 & $-$0.117 & $-$0.602$^{*}$ & 0.334$^{***}$ & $-$0.109 & 0.051 \\ 
  & (0.263) & (0.071) & (0.302) & (0.082) & (0.275) & (0.075) \\ 
  & & & & & & \\ 
 june & $-$0.008 & $-$0.066 & $-$0.194 & 0.315$^{***}$ & $-$0.094 & $-$0.026 \\ 
  & (0.256) & (0.070) & (0.291) & (0.081) & (0.258) & (0.074) \\ 
  & & & & & & \\ 
 july & 0.335 & $-$0.060 & $-$0.447 & 0.155 & $-$0.742$^{*}$ & 0.046 \\ 
  & (0.282) & (0.071) & (0.306) & (0.082) & (0.304) & (0.075) \\ 
  & & & & & & \\ 
 dm1 & 0.307 & $-$0.092$^{*}$ & $-$0.138 & $-$0.097 & 0.016 & $-$0.129$^{**}$ \\ 
  & (0.175) & (0.046) & (0.194) & (0.053) & (0.184) & (0.048) \\ 
  & & & & & & \\ 
 dm2 & 0.006 & $-$0.003$^{*}$ & $-$0.005 & $-$0.002 & 0.011 & $-$0.002 \\ 
  & (0.005) & (0.001) & (0.006) & (0.002) & (0.006) & (0.001) \\ 
  & & & & & & \\ 
 dm3\_edu & 0.051 & 0.038 & 0.184 & $-$0.005 & $-$0.173 & $-$0.143$^{***}$ \\ 
  & (0.139) & (0.039) & (0.156) & (0.045) & (0.152) & (0.042) \\ 
  & & & & & & \\ 
 unempl & 0.376 & $-$0.026 & $-$0.067 & $-$0.076 & 0.332 & 0.058 \\ 
  & (0.209) & (0.053) & (0.229) & (0.061) & (0.224) & (0.055) \\ 
  & & & & & & \\ 
 married & 0.095 & 0.019 & 0.053 & 0.133$^{*}$ & $-$0.204 & 0.098$^{*}$ \\ 
  & (0.169) & (0.046) & (0.188) & (0.054) & (0.180) & (0.049) \\ 
  & & & & & & \\ 
 Constant &  &  & 0.593 & $-$0.074 &  &  \\ 
  &  &  & (0.712) & (0.180) &  &  \\ 
  & & & & & & \\ 
\hline \\[-1.8ex] 
Observations & 212 & 2,396 & 212 & 2,396 & 212 & 2,396 \\ 
Log Likelihood &  &  & $-$137.201 & $-$1,637.219 &  &  \\ 
Akaike Inf. Crit. &  &  & 298.401 & 3,298.439 &  &  \\ 
\hline 
\hline \\[-1.8ex] 
\textit{Note:}  & \multicolumn{6}{r}{$^{*}$p$<$0.05; $^{**}$p$<$0.01; $^{***}$p$<$0.001} \\ 
\end{tabular} 
\end{table} 
